Week 10 Problem Session: ICM workshop

Running the recursive ICM calculation on a payout structure and stack distribution nobody has seen, then a two-player sanity check by hand

26 April 2027

Before the session#

Read the lecture page. Bring the $417.86 / $325.00 / $257.14 flagship $EV figures and the naive $500 / $300 / $200 comparison — the session runs the same calculation on a different stack distribution and payout structure to see if the same pattern (chip leader loses value, short stack gains it) still holds.

In the session#

The group works through the recursive calculation by hand for a new four-player final table with a top-heavy payout structure, computing each player’s placement probabilities one place at a time before converting to $EV — checking along the way that every row and every column of the placement matrix still sums to 1.

Second half: a two-player special case, where ICM collapses to a simple linear formula (P(1st) for the chip leader, weighted against the two payouts) — confirming that the general recursive method and the direct two-player formula agree, as a sanity check on the general method rather than a new result.

Afterwards#

Nothing is marked this week. Going into week 11: the difference between “chip EV,” “tournament dollar EV” (this week), and a bankroll that has to survive many such tournaments — the question week 11 actually asks.